Casual, consensual sexual encounters among college students have been identified as the hookup culture, which is prevalent on college campuses (Napper, Montes, Kenney, & LaBrie, 2015). Although unsafe sexual behavior impacts college students on campuses across the world (Bersamin, Paschall, Saltz, & Zamboanga, 2012; Brown-Rice & Furr, 2015; Napper et al., 2015) a number of factors including unsafe sexual behavior, unwanted pregnancies, transmission of STIs/STDs, and inadequate knowledge about campus risks could potentially impact the health of college students (LaBrie et al., 2014). Although traditional messages communicated from parents have primarily focused on the importance of waiting to have sex until marriage, Moilanen and Raffaelli (2010) stated that the messages shared with the rising generation regarding the risks associated with sexual engagement are centered around unwanted pregnancy and STDs. Although abstinence is still a message shared from parents to their children, familistic values are used to promote awareness, personal responsibility, and the importance of decision- making (Manago et al., 2015). Manago et al. (2014) identified a cultural difference
between the degrees of family interdependence versus individual independence. Familism in a traditional culture focuses on practices and values, which are strong characteristics of family interconnectedness (Manago et al,, 2014; Manago et al., 2015). Although familism places emphasis on positive family relationships, one struggle for college students is the separation from family after moving to a college/university (DiBello et al., 2015).
DiBello et al. (2015) identified that having a positive family relationship and enriched values that have been taught throughout a child’s upbringing is often challenged
when leaving a stable environment to attend a college. Familism is not only important within the Hispanic community, but is also embraced by people of other ethnicities (Steidel & Contreras, 2003).Researchers have demonstrated that familism is used to defend against risky behaviors and that women tend to participate in risky behaviors less than men (Bersamin et al., 2012; DiBello et al., 2015). In this chapter, I discuss the research design and rationale, methodology, population, sampling procedures, procedures for recruitment, data collection, instrumentation and operationalization of constructs, data analysis plan, threats to validity, and ethical procedures.
Research Design and Rationale
To examine the relationship between the independent and dependent variables of this study, I used a quantitative correlational design. The independent variable was familism level and the dependent variable was self-reported sexual behavior.
Demographic variables were included as covariates, which included age, gender, marital status, place of residence, socioeconomic status, religion, number of sex partners, STI history, and use of protective contraception. I examined the views of familism for each college student to determine whether the cultural value of familism impacted how students make decisions regarding sexual activities.
A correlational study design is used to determine whether the variables selected for the study are related to one another (Creswell, 2008). Correlational research allows researchers to collect significantly more data than when conducting an experiment (Creswell, 2008). Although correlational research usually occurs outside of a lab, results are usually more applicable to individuals’ lifestyles. Another benefit of correlational
research is that it opens up further research for scholars to examine the phenomenon or relationship of the variables in an in-depth approach. For researchers to determine the strength and direction of the relationship between variables, correlational research is used to obtain the findings of the study so scholars can take an in-depth look at causation. Although there are advantages of using a correlational study, correlational research is limited to only uncovering the relationship between variables and cannot provide a conclusive reason for why there is a relationship or establish causation between variables (Creswell, 2009).
Creswell (2009) argued that correlational research is unable to reveal which variables influence the others. The correlational design was used to study whether an increase or decrease in the independent variable familism predicted an increase or decrease in the dependent variable sexual behavior. Correlational studies can be
conducted by using naturalistic observation, survey method, or archival research (Losada et al., 2010). If a researcher is using naturalistic observation, he or she needs to view the variable of interest in its natural environment without manipulation. In some cases, this collection of data could provide researchers with the opportunity to further the research for the selected variable. The downfall of using this method is that it could be time consuming and expensive to conduct this type of research. Selecting the naturalistic method would require the researcher to have sufficient funding to conduct this type of study.
Survey methods are usually inexpensive and provide researchers with the ability to gather large amounts of data in a short period of time (Losada et al., 2010). The survey
method is more flexible than naturalistic observation and archival research. For researchers who decide to employ this method, one constraint to consider is that participants can affect the outcome of the study (Huebner & Perry, 2015). Issues that generally arise with the survey method include participants trying to please the
researcher, lying to make themselves look better, or having inaccurate memories about the questions that are asked through the survey (Huebner & Perry, 2015).
If the researcher decides to use archival research, large amounts of data can be collected, which reduces the struggle of collecting data for the study (Campbell & Stanley, 1963). Although this provides the researcher with enormous data, the research has no control over how the data was collected, and previous research may be unreliable. The time constraints that could have impacted how data were collected included when students are in attendance at the selected college (see Campbell & Stanley, 1963).
Sayegh and Knight (2011) stated that correlational research allows researchers to collect more data and conduct more in-depth studies to examine the relationships between variables. These authors conducted a study using a correlational design and found that this allows researchers to study phenomena or a relationship. Although familism has been studied in-depth in the Hispanic population, Schwartz (2007) indicated that future
research should address how familism impacts other cultures. There was limited research about how familistic views impact decisions college students make regarding sexual behavior (Wentland & Reissing, 2011), and the current study was conducted to investigate the relationship between familism and sexual behavior for students on a college campus.
Methodology Population
The participants for this study were drawn from college students age 18 and older attending Walden University. The enrollment of Walden University is more than 52,600 students from all 50 U.S. states and more than 155 countries. Students who are enrolled at Walden University are pursuing bachelor’s, master’s, and doctoral degrees in various disciplines (i.e., health sciences, criminal justice, nursing, public health, counseling, and other disciplines). The purpose for choosing this university was the diverse population of students from different cultural backgrounds and geographical locations. To gain
permission from Walden University to use the Walden Participant Pool, I had the
institutional review board (IRB) reviewed the research design and process for this study. Sampling Strategy
The sampling strategy that was used for this study is a convenience sampling strategy. A convenience sampling strategy is a method of drawing representative data by selecting participants due to the ease of volunteering, availability, or easy access to participate (Leiner, 2014). The advantage of using this type of sampling is the availability and speediness with how data can be gathered. The disadvantage of using this type of sampling is that the sample may pose potential problems for not representing the
population as a whole and may cause volunteers to be biased about there responses. The eligible criteria that was used for this convenience sample is that participants must be over the age of 18 years and a student at Walden University. If these participants are a
student with Walden University, then they must setup an account with the Walden participant pool in order to complete this survey.
Power Analysis
In order to determine the appropriate sample size for this study a power analysis was conducted. An alpha level of .05 and power of .80 was used to calculate the sample size. Since I utilized more than one instrument for this study, the goodness-of-fit-tests along with the A priori was used to compute the required sample size needed for the sample population. According to Danube, Vesico, and Davis (2014), the G-Power
analysis can be used to compute the effect size, alpha, and power to determine the sample size needed for the study.
Sample Size
I utilized G-Power to calculate the sample size for my study as described below using an A Priori calculation for a linear multiple regression determine how many participants would be needed for this statistical test. After completing the power analysis calculations, it was determined that 118 participants are needed for the sample population in order to conduct the multiple regressions. Based on the computation of the goodness- of-fit-test, the power (0.80), alpha (0.05), medium effect size, which is (0.15) which is usually the medium effect size is the norm (Danube, Vesico, & Davis, 2014), and number of predictors (10) were calculated using G*Power to determine the actual sample size required for this study, which is 118 participants. Since the sample size is not
significantly larger than 120, which would be required for a multiple regression, it would be appropriate to use the sample size of 118 so that the study would have enough data to
determine whether to accept or reject the null hypothesis. Another step in this process would be to gather gender distribution, which would require having half male and half female participants that take the survey.
Procedures for Recruitment, Participation, and Data Collection
Recruitment. In order to participate in this study, the Walden Participant Pool is a virtual bulletin board that provides researchers with the opportunity to connect
researchers with participants. By posting this research study on the Participant Pool, active members in the Walden community have the opportunity to participate in this research study. While this is a great resource for researchers, participants who are members of the Walden Participant Pool are made up of diverse students and faculty members that could potentially participate in this study.
Participation. In order to participate in this study, participants are required to be a Walden University student. Participants who also meet the criteria of being over 18 will be provided an explanation of the study on the Walden Participant Pool website. Since this study will not compensate participants, each participant will be informed about his or her participation and that this study is completely voluntary and that all information collected will be stored on a secure server and remain confidential (participant names will not be collected). Regarding informed consent, it has been determined that the easiest method for providing informed consent to students in this process would be to include this as the first page of the survey and participating in the survey would indicate informed consent. Participants would be informed through this process that their participation is
voluntary, will not impact their relationship with the university, and that they can stop participating at any time.
Data collection. The data collection method for this correlational survey used informed consent, a demographic survey, as well as closed-ended questions, and a self- administered online survey. The questionnaire was distributed through Survey Monkey, where all information will be submitted and stored upon completion by each participant of the survey. Albaum and Smith (2006) defined that Survey Monkey is a web-based software that allows user the ability to create surveys or generate reports without having prior knowledge or experience with programming. Survey Monkey enables the user to create a survey in an online environment, which can provide feedback with a
downloadable report to examine findings from the survey (Albaum & Smith, 2006). Typically, this web-based software program is used to aid researchers, report results, or to create polls (Albaum & Smith, 2006). Albaum & Smith (2006) state that Survey Monkey is a secure web based program that is encrypted with user-id and password protection, which will only allow individuals that are given access to complete the survey online. This web-based program can be used on PC or MAC computers, which allows flexibility with participants who may be operating these types of devices.
The online survey was chosen as a way to capture confidentiality so information will not have to be stored in a file cabinet. Because this research study stored all
information on a secure site, this eliminated the risk for any information to be comprised. According to Yu et al. (2015) a study was conducted by these researchers by recruiting participants to take an online survey about the risk of men having sex with men (MSM)
and the risk associated with contracting HIV. In order to keep results confidential, the researchers decided to conduct an online survey to see if they could receive more input from MSM and receive positive feedback from administering this survey online. The purpose of this study was to examine the relationship between familism and sexual behavior among college students. In this process, students that are a part of the Walden Participant community tool will be sent the survey directly to the school email account along with a detailed description of the purpose of the survey. Participants who decide to complete and submit the survey online will provide informed consent by submitting the online survey through Survey Monkey. Once all participants have completed the survey, this information will be stored on a secure website that keeps all information stored.
Instrumentation and Operationalization of Constructs Instruments
Instruments were chosen to quantitatively measure the independent variable familism and the dependent variable sexual attitudes to determine the relationship between both variables (Lugo Steidel & Contreras, 2003; Schwartz et al., 2007). Although previous researchers have focused on the Latino population, the same instruments will be used to measure and assess participants in this study. According to Creswell (2009), the researcher must provide information about the research that is being utilized for the study. By using a published tool, the researcher must receive approval from the person(s) that created the instrument before moving forward to conduct the study.
Demographic questionnaire. A sociodemographic questionnaire was included in this study to obtain background information for each participant in the study. The items that have been included in this questionnaire are age, grade level classification, gender, race/ethnicity, religious affiliation, and country of origin. The demographic questionnaire for this study was used from a previous study, which will use these listed variables (Table 1) to decipher between each participant of the study to examine the relationship between familism and sexual behavior. The variables for this study were coded as the following (i.e., age-age, Education Level-EDU, gender-GEN, race/ethnicity-ETHN, religious affiliation-REL, number of sex partners- SEX Part, STI history- STI Hist, protective contraception- PC).
Measure of Attitudinal Familism Scale. The Attitudinal Familism Scale (Lugo Steidel & Contreras, 2003) will be used to assess familism in this study (see Appendix C for permission letter). This scale consists of 18-items assessing familial support, familial interconnectedness, familial honor, and subjugation of self to family (Lugo Steidel & Contreres, 2003). The instrument uses 10-point Likert scale that ranges from one (strongly disagree) to 10 (strongly agree) to measure each of the 18-items in the Attitudinal Familism Scale.
According to Lugo, Steidel, and Contreras (2003) this instrument was developed and validated using Hispanic adults in Cleveland, Ohio. Since the validation of this instrument was originally found reliable and valid when used with Hispanics, the Cronbach’s alpha was used for internal consistency, confirmatory factor analysis for construct validation, and internal consistency reliability assessment for stability. Since the
population differs from the original population, a retest for the selected population for this study was conducted to determine reliability and validity compared to the original overall score, which ranges from .70 to .80. By using this survey, one will determine if the same measures within the structure of familism is consistent with other acculturation levels, ages, and ethnicities. Regarding reliability, Lugo, Steidel, and Contreras (2003) identified Cronbach’s alpha for the factors were .80 for the overall score. According to Lugo, Steidel, and Contreras (2003) Familial Support (.72), Familial Interconnectedness (.69), Familial Honor (.68), and Subjugation of Self for Family (.56) were subscales that were inter-correlated by computing the overall mean for the whole scale. The results determined that a higher score suggested a higher endorsement of familism. In order to test validity of the scale, a correlation was conducted between all familism scores, which would determine whether that is a positive or negative correlation between linear
acculturation scores and overall familism (Lugo Steidel & Contreras, 2003). The results of the test for validity found that there was a significant negative correlation between linear acculturation score and overall familism (Lugo Steidel & Contreras, 2003).
Measure of the Brief Sexual Attitudes Scale. The Brief Sexual Attitudes Scale (Hendrick, Hendrick, & Reich, 2006) will be used to assess attitudes about sex in this study. Since this instrument is listed in the Measurement Instrument Database for the Social Sciences (MIDDS) permission was not necessary as the Institute for Business (administrator) grants privilege for anyone to use instruments published on this site. This scale consists of 23-items rated on a five-point likert scale that ranges from strongly agree to strongly disagree. Currently, the BSAS scale breaks the score into four subscales (i.e.,
Permissiveness, Birth Control, Communion, and Instrumentality), which Cronbach’s alpha is listed below in each of the subgroups, which range from 1.0 to 5.0. (Lower score indicates a greater amount towards that attitude) The permissiveness subscale measures an individual’s attitude that is open to relationships regarding sex (Hendrick, Hendrick, & Reich, 2006). The Birth Control subscale measures an individual’s attitude to be
responsible for providing contraception to protect against pregnancy (Hendrick,
Hendrick, & Reich, 2006). The Communion subscale measures an individual’s attitude toward the importance of melting together with their sex partner (Hendrick, Hendrick, & Reich, 2006). The Instrumentality subscale measures an individual’s attitude toward enjoying the physical sex (Hendrick, Hendrick, & Reich, 2006).
According to Hendrick, Hendrick, and Reich (2006), this instrument was
developed and validated to assess multi-dimensional attitudes toward sex. These authors conducted reliability and validity analyses using a population of 79 undergraduate students from a human sexuality class from a large southwestern university. The
Cronbach’s alpha for the overall score is .80 and the subscales were Permissiveness =.95; Birth Control =.88; Communion =.73; and Instrumentality =.77. In order to determine validity and reliability, internal consistency reliability will be calculated with Cronbach’s alpha with the sample to determine if they match, are close or are better than the original subscale scores. Since this scale had to be revalidated from its original 43-item
instrument, this 23-item instrument was validated and results indicated better
psychometric properties than the longer Sexual Attitudes Scale (Hendrick, Hendrick, & Reich, 2006). Based on the subscale intercorrelations and correlations for BSAS,
Hendrick, Hendrick, and Reich (2006) found that the lack of racial, ethnic, and
socioeconomic diversity will be needed for further research with this instrument. For the subscale item birth control, the alpha for this subscale was excellent, but the correlation of .57 reflects inconsistencies among college students use of birth control and their attitudes about it.
Operationalization
Table 1
Demographic Variables
Variable Coding Value
Age Age 0=18-up
Education Level EDU 0=Highschool/GED
1=Undergraduate 2=Masters 3=Doctoral/PhD
Gender GEN 0= Male 1=Female
Ethnicity ETHN 0= White 1=Black or African
American 2=Hispanic or Latino 3=Native Hawaiian or Other Pacific Islander 4= American Indian or Alaskan Native 5= Asian 6=Two or more races 7=Unknown